The zebrafish is a powerful vertebrate genetic model and has recently been used as a system to understand cancer. We have isolated 19 cell cycle mutants and several mutated genes are associated with cancer in our model system. Our studies of B-myb identified a novel pathway that affects cyclin B and participates in cancer, and our recent study of a separase mutant demonstrated that genome instability predisposes epithelial cells to cancer. For this proposal, we plan to investigate two independent cancer models based on ras pathway activation. RAS family members and the downstream targets are mutated in a number of human cancers, including melanoma and rhabdomyosarcoma. In the first model, BRAF and p53 deficiency induced melanoma. In a second model, a constitutive KRAS allele leads to embryonal rhabdomyosarcoma at 10 days of age. This is the fastest tumor model in any vertebrate. In both cases, we have demonstrated that the cells that initiate the tumor are transplantable and represent early stages of melanocytic and myogenic differentiation, respectively. We hypothesize that RAS induction will activate similar self-renewal pathways within tumors of different tissue types. Using transgenic fish, we plan to label the cancer cells with a number of reporter constructs and determine the subpopulation of tumor cells that are transplantable. The models will be further used to understand the signaling pathways operative in the cancer-initiating cell population. We plan to determine if the wnt-prostaglandin pathway can modify the ability of the cancer-initiating cell to function. We also plan to undertake a chemical screen defining molecules that will inhibit cancer stem cells. Two novel chemicals have already been found that block an expansion of early neural crest progenitors in the BRAF;p53 deficiency embryos. These chemicals are now being tested in our zebrafish model for the ability to block adult melanoma. Our studies will lead to a better understanding of cell heterogeneity within a cancer and will also lead to potential therapies for rhabdomyosarcoma and melanoma. PUBLIC HEALTH RELEVANCE: Cancer is a difficult disease to treat because cells remaining after chemotherapy can recapitulate the tumor, leading to relapse. Here we plan to examine the differentiation of these rare cells within a tumor, and find pathways that disrupt them. By defining such pathways, new therapies can be designed to use in conjunction with conventional chemotherapy in the treatment of cancer.